The paper introduces PARED, an approach that uses inverse reinforcement learning to derive an explicit, interpretable reward from expert demonstrations. By training a lightweight discriminator on response-level features, PARED can guide both inference-time reranking and on-policy reinforcement learning, improving alignment for models that have not undergone supervised fine-tuning and further refining those that have. The method also supports contextual alignment, enabling a single policy to adapt its responses to different audience preferences.
By Micha{\l} Wili\'nski, Liu Leqi, Chirag Nagpal
The paper proposes a method for predicting long‑term behavior in chaotic systems by learning a time‑invariant evolution operator that can jump across large time spans in a single evaluation. It derives consistency equations that the operator must satisfy, incorporating differential and compositional objectives tied to physical dynamics. Experiments on five chaotic systems show that this physics‑distilled consistency approach yields more accurate short‑term trajectories and long‑term statistics, survives temporal extrapolation, and requires only one‑tenth the evaluations of autoregressive rollouts.
By Lufang Chiang, Jiachen Yao, Thomas Y. L. Lin, Anima Anandkumar
SAGE, a 105‑million‑parameter semantic audio generative encoder, compresses music waveforms into compact latent representations by distilling embeddings from a pretrained audio‑text model. It achieves a three‑way balance—high reconstruction quality, rich semantic latent structure, and fast inference—running three times faster than Stable Audio Open while matching or surpassing larger models in listening‑test quality and objective metrics. Across all nineteen latent‑semantics probing tasks, SAGE sets new state‑of‑the‑art performance both within and outside its training domain.
By Francesco Brigante, Luca Cerovaz, Davide Marincione, Giorgio Strano, Luca Zhou, Emanuele Rodol\`a, Michele Mancusi
AOV-GS is an active 3D Gaussian mapping system that uses frozen foundation models (SAM and CLIP) to guide exploration toward semantically informative regions. During coarse exploration, it re‑ranks geometrically promising views based on CLIP mask‑embedding novelty from SAM instance segmentation, biasing the robot’s trajectory. The system fits a lightweight scene‑specific language field offline to the collected CLIP embeddings, enabling open‑vocabulary text‑query segmentation on novel views and improving language‑field mIoU by 1.3–1.8 percentage points over geometry‑only planning on Replica and Matterport3D.
By Rodion Shkokov, Tatiana Zemskova, Dmitry Yudin
arXiv:2610.30016v1 Announce Type: cross
Abstract: Understanding large, complex, and rapidly evolving code bases remains a core challenge for software engineering at scale, hindering onboarding, maint...
By Sujoy Roychowdhury, Joseph H. R. Isaac, Jiangning Gao, Martin Klitte, Diarmuid Corcoran
arXiv:2610.29424v1 Announce Type: new
Abstract: Among existing embedding compression approaches, low-rank factorization provides a simple and mathematically principled solution by approximating the e...
By Oliver Giudice, Anna Russo Russo
arXiv:2610.29645v1 Announce Type: new
Abstract: Large language models (LLMs) have been explored as surrogates for human survey respondents. A more promising application is data augmentation, where LL...
By Esko Lehtonen, Tommi Buder-Gr\"ondahl
The paper evaluates whether recent LLM-based node classification methods on text‑attributed graphs truly outperform simple baselines. By tuning four baseline models on the same data splits and comparing them to the best reported LLM results, the authors find that while some gains are modest (e.g., Correct & Smooth trailing by 2.3–2.8 percentage points), several reported improvements remain significant. They also highlight evaluation risks such as test‑set hyperparameter tuning, misuse of test‑pool labels, undisclosed baseline reuse, and inconsistencies across papers.
By Michal Podstawski
SCARCE is a label‑free representation‑learning framework that improves graph‑level out‑of‑distribution generalization. It uses an environment partition of training graphs—derived from metadata or self‑supervised clustering—to warm‑start a variational graph autoencoder and then applies an adversarial confusion mechanism that prevents a discriminator from distinguishing environments while preserving reconstruction fidelity. Theoretical analysis shows that this environment confusion contracts environment‑discriminative directions, effectively reweighting features to reduce reliance on spurious correlations, and experiments on synthetic and DrugOOD datasets demonstrate state‑of‑the‑art OOD performance among label‑free methods.
By Tianjun Yao, Chen Xi, Xinyang Li, Lixian Zhong
The paper introduces LAW (Loopholing All the Way), a continuous generative model that relies entirely on Self‑Conditioning (SC) to propagate expected embeddings across diffusion steps, decoding tokens only once at the end. LAW outperforms both discrete and continuous diffusion models on Sudoku, Maze, and TinyGSM benchmarks, achieving higher accuracy and solving a larger proportion of GSM8K problems. Fine‑tuning a pre‑trained Uniform Diffusion Model with LAW significantly boosts its performance, and on OpenWebText LAW approaches autoregressive models while maintaining a favorable Pareto frontier.
By Adrien Deschenaux, Nadezhda Ilieva, Justin Deschenaux, Manuel Madeira, Dorina Thanou, Aur\'elien Lucchi, Antonio Orvieto, Caglar Gulcehre, Pascal Frossard
arXiv:2610.29688v1 Announce Type: cross
Abstract: Large pretrained models such as CLIP are widely used as frozen feature extractors. However, their embeddings inherit spurious correlations from train...
By Divyaansh Mertia, Chirag Bhutra, Himanshu Shekhar, Shubham Bagwari, Pratik Mazumder
arXiv:2509.03242v3 Announce Type: replace
Abstract: Testing Deep Learning (DL)-based systems is an open challenge. Although it is relatively easy to find inputs that cause a DL model to misbehave, th...
By Gianmarco De Vita, Nargiz Humbatova, Paolo Tonella
arXiv:2610.29210v1 Announce Type: new
Abstract: Speculative decoding uses a target model to verify a tree of draft tokens in parallel. We study whether the resulting next-token distributions can help...
By Jungseob Lee, Giyoon Jang, Chailin Nam, Jiwoo Kum, Sihyeon Lee, Kyojun Choo, Jihun Song, Eunseo Song, Chanjun Park
arXiv:2610.25192v2 Announce Type: replace
Abstract: Text processing is crucial for low-resource languages as it enables the development of tools and technologies that bridge the digital divide and pr...
By Shantipriya Parida, Debasish Dhal, Sonal Khosla, Sambit Sekhar, Muhammed Abdur Rahmaan Kamaldeen
arXiv:2610.29651v1 Announce Type: new
Abstract: Text-to-Image models reproduce societal stereotypes, and existing mitigation applies one demographic correction to the whole scene. In multi-person sce...
By Venkatesh Thirugnana Sambandham, Torsten Schon
arXiv:2610.29811v1 Announce Type: new
Abstract: Compositional zero-shot learning (CZSL) aims to recognize novel combinations of previously observed objects and attributes. Prior work showed that larg...
By Puru Vaish, Meelad Dashti, Christoph Brune, Pascal Mettes, Tatiana Tommasi, Nicola Strisciuglio
arXiv:2610.29866v1 Announce Type: new
Abstract: Geometric structure is essential for robotic manipulation and understanding the 3D world. Most VLA models rely on 2D representations of current observa...
By Qinfan Zhang, Yifan Luo, Haoyuan Shi, Guanlin Liu, Han Dong, Haozhe Shan, Yong Dai, Xiangcong Ren, Xiaozhu Ju
arXiv:2610.30023v1 Announce Type: new
Abstract: This paper addresses the problem of creating immersive volumetric videos from sparse and low-resolution capture resources. Conventional dense multi-vie...
By Chenxu Zhou, Zhiyuan Yu, Yudong Jin, Yinji Shentu, Yu Zhang, Jiaming Sun, Xiaowei Zhou
arXiv:2610.29246v1 Announce Type: cross
Abstract: Predictive representations for robot manipulation should connect scene geometry with the state changes induced by actions. Our intuition is that unde...
By Xiujian Liang, Liuyuan Na, Zehao Du, Jiacheng Xu, Weixi Song, Mingyang Sun, Xingjun Ma, Cewu Lu, Jianhua Sun
arXiv:2602.09850v3 Announce Type: replace
Abstract: Industrial anomaly detection demands precise reasoning over fine-grained defect patterns. However, existing multimodal large language models (MLLMs...
By Peng Chen, Wenqiang Wang, Yunkang Cao, Chengliang Liu, Wei Wang, Mingbo Yang, Li Shen, Wenqi Ren, Xiaochun Cao, Chao Huang